Beta cells of the pancreas, which make and secrete insulin, do not respond like those of non-diabetic subjects when type 2 diabetes is present. Specifically, subjects suffering from type 2 diabetes have a blunted or even absolute loss of first phase and a severely blunted second phase insulin release in response to glucose. In conjunction with this, and despite all treatments currently available to treat diabetes, beta cell function continues to deteriorate over time. With the data now available from the United Kingdom Prospective Diabetes Study (Sept. 1998) this point was brought home even more forcefully. Despite continual monitoring of patients enrolled in the study, euglycemia could not be maintained even with intensive therapy, because of declining beta cell function. We have been working for some time with GLP-1, a naturally occurring peptide produced and released from the gut in response to food. The amount released depends on the amount of glucose and fat that has been ingested. After its plasma levels increase, GLP-1 binds to the GLP-1 receptor (GLP-1R) on beta cells, and increases PKA activity because of cAMP generation. Downstream of the increased PKA activity, insulin release occurs also in a glucose-dependent manner, plasma glucose having risen because of the meal. The end result is a restoration of plasma glucose back to baseline. Consequently, GLP-1 analogs and GLP-1R agonists are under intense study as treatments for type 2 diabetes. A naturally occurring GLP-1R agonist, exendin-4, is now available for treatment. It binds to GLP-1R with ten times more avidity than does native GLP-1. It contains many amino acids homologous to GLP-1 and, interestingly, contains a unique 9 amino acid C-terminus. We have extensively investigated the functional significance of the C-terminus by studying the effects on GLP-1 when it is added to the C-terminus of GLP-1 and by deleting the 9 amino acids sequentially and in triads from exendin-4. Its addition to GLP-1 improved its binding affinity to its own receptor and its deletion from exendin-4 abrogated its increased affinity for GLP-1R. Therefore, for superagonist properties of exendin-4 to GLP-1R the 9 amino acid C-terminus is required. More recently, we have been working on a GLP-1 fusion protein whereby GLP-1 is fused to human transferrin, thus extending its half-life, in vitro, by several days. The fusion protein is synthesised in yeast by recombinant technology and secreted into the yeast broth from which it is extracted. A major impediment to using GLP-1 itself is that its half-life is only a few minutes because it is rapidly inactivated by removel of its first two amino acids at its N-terminus by dipeptidyl peptidase 1V. We have modified the GLP-1 in the fusion protein so that it is not a substrate for dipeptidyl peptidase 1V. We have carried out in vivo work in small and large animals to delineate kinetics and dosing schedules of the fusion protein and FDA approval has been granted for a dosing schedule in humans. Additionally, we have now fused exendin-4 to transferrin and found there is no loss of exendin-4 activity in rodents, when compared to exendin-4 alone. We hope to receive FDA approval for its study in humans. [unreadable] Another insulinotropic peptide, GIP, is also released from the gut after eating, but, based on BLSA data, beta cells of the pancreas appear resistant to its effects once blood glucose exceed 126 mg/dl. Additionally, in vivo experimental protocols in humans using exogenous GIP have shown that beta cells do not increase insulin secretion in response to exogenous human GIP. GIP is also subject to rapid breakdown by dipeptidyl peptidase 1V. Therefore, we are currently testing in type 2 diabetic humans if a dipeptidyl peptidase 1V-resistant GIP peptide might be successful at increasing insulin secretion from beta cells. This study involves 24 subjects to which increasing concentrations of the peptide have been given up to a maximum of 20 ng/kg/min. This protocol will be finished this year.